340 GORDON E. DICKERSON 



son, 1949) from the average ratio per locus of maximum potential superiority 

 of hybrid over average of all possible Fi's to the observable mean advantage 

 of the Fi's over the inbred lines themselves (Fig. 21.3). Using (1 — q) for 

 frequency of the less favored allele, /for Wright's inbreeding coefficient, k for 

 degree of dominance, as before, and d = {A A — aa)/2: 



H = Maximum heterozygote =C -\- {l -\- k) d 



Fi = Mean of Fi crosses = C-{-2q[\ + k{l-q)]d 



P = Mean of inbred lines = C -\-2q [\ -j- k (1 - q) {I - f) ] d 



(H-Fi) {k-l)[l-2q(l-q)]-h2{l-q)^~ 



Hence , 



Fi-P 2kjq{\-q) 



Underpartialor complete dominance, equilibrium (1 — 9) = {k —\)/2k = 

 0, except for reverse mutation pressure. When the parental lines are homozy- 

 gous (/ = 1), mean (1 — q) lies between .05 and .1 and mean inbred yield is 

 about 40 per cent of that for Fi crosses, the maximum increase of hybrid over 

 average Fi would lie between 3 and 7 per cent. There is little reason to sup- 

 pose that present better hybrids approach the maximum. The potential maxi- 

 mum increase over open-pollinated varieties increases rapidly with degree of 

 heterozygote advantage {k), approaching {Fi — P) or about 50 per cent in 

 corn yield. 



The evidence thus far obtained in swine suggests that the genetic basis of 

 variation in net productivity is fundamentally quite similar to that in corn. 

 This indicates challenging possibilities for increasing productivity of swine 

 by utilizing potential heterosis. 



Heterozygote Advantage for Single Loci and Chromosome 

 Segments in Other Species 



Dobzhansky (1949) has shown experimentally that natural selection 

 favors individuals heterozygous for inversion chromosome segments in 

 Drosophila. He also has shown that the resulting equilibrium between fre- 

 quency of alternative homologous segments fluctuates with locality and sea- 

 son of the year, depending on relative selective advantage of alternative 

 "homozygous" segments. He postulates natural selection for increased co- 

 adaptation between alternative segments in heterozygotes within each inter- 

 breeding population. Demonstration of heterozygote advantage at individual 

 loci would be difficult, since any one locus usually would account for only a 

 small part of the total variability in selective value or in any complex charac- 

 ter. However, some cases of presumably single gene mutations exhibiting 

 heterozygote advantage have been reported (Jones, 1945; Gustafsson, 1946, 

 1947). The "yellow" gene of the agouti series in mice (Danforth, 1927; 

 Dickerson and Gowen, 1947) provides a classic example of manifold effects 



